CN114283935B - Pain grade assessment method and system - Google Patents
Pain grade assessment method and system Download PDFInfo
- Publication number
- CN114283935B CN114283935B CN202111582180.5A CN202111582180A CN114283935B CN 114283935 B CN114283935 B CN 114283935B CN 202111582180 A CN202111582180 A CN 202111582180A CN 114283935 B CN114283935 B CN 114283935B
- Authority
- CN
- China
- Prior art keywords
- brain network
- brain
- pain
- network
- coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 208000002193 Pain Diseases 0.000 title claims abstract description 126
- 230000036407 pain Effects 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 35
- 210000004556 brain Anatomy 0.000 claims abstract description 310
- 239000000523 sample Substances 0.000 claims abstract description 63
- 230000037324 pain perception Effects 0.000 claims abstract description 34
- 239000013068 control sample Substances 0.000 claims abstract description 28
- 239000003550 marker Substances 0.000 claims abstract description 24
- 238000004220 aggregation Methods 0.000 claims description 72
- 230000002776 aggregation Effects 0.000 claims description 46
- 238000012417 linear regression Methods 0.000 claims description 29
- 230000000763 evoking effect Effects 0.000 claims description 25
- 208000000114 Pain Threshold Diseases 0.000 claims description 15
- 230000037040 pain threshold Effects 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 5
- 238000012935 Averaging Methods 0.000 claims description 3
- 108010015046 cell aggregation factors Proteins 0.000 claims description 3
- 208000000094 Chronic Pain Diseases 0.000 abstract description 17
- 238000011156 evaluation Methods 0.000 abstract description 11
- 230000008447 perception Effects 0.000 abstract description 5
- 230000006698 induction Effects 0.000 abstract description 3
- 230000001575 pathological effect Effects 0.000 abstract description 3
- 230000008914 somatic perception Effects 0.000 abstract description 2
- 238000002599 functional magnetic resonance imaging Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 210000001638 cerebellum Anatomy 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 210000004884 grey matter Anatomy 0.000 description 2
- 230000001939 inductive effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000284 resting effect Effects 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 239000000090 biomarker Substances 0.000 description 1
- 230000007177 brain activity Effects 0.000 description 1
- 230000003925 brain function Effects 0.000 description 1
- 208000029028 brain injury Diseases 0.000 description 1
- 210000003710 cerebral cortex Anatomy 0.000 description 1
- 210000001175 cerebrospinal fluid Anatomy 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000002582 magnetoencephalography Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007634 remodeling Methods 0.000 description 1
- 230000002739 subcortical effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 210000004885 white matter Anatomy 0.000 description 1
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The application provides a pain grade evaluation method and a system, wherein a brain network corresponding to each sample data is established through sample data of a pain sample, a control sample and an induction sample, and topology attribute values of each brain network model are calculated to obtain dynamic variation gradient attributes of the brain network corresponding to the pain sample data relative to the brain network corresponding to the control sample data and weight values of each brain region participating in pain perception, so that a dynamic pain marker is obtained, and the perception degree of the brain of a chronic pain patient can be accurately and quantitatively evaluated according to the dynamic pain marker. According to the method and the system provided by the application, variability of the pain sample compared with the control sample can be intuitively reflected through the dynamic variation gradient attribute of the brain region, pathological influence of chronic pain can be effectively removed through the obtained weight value of each brain region participating in pain perception, and the obtained dynamic pain marker is focused on the somatic perception layer, so that a more objective pain grade evaluation result is obtained.
Description
Technical Field
The application relates to the technical field of brain science, in particular to a pain level assessment method and a pain level assessment system.
Background
Chronic pain is the most common and important symptom in a disease, and as a subjective symptom, the degree of pain is generally evaluated based on the patient's own feeling. Therefore, it is difficult to quantify the pain level, and how to objectively evaluate the pain intensity has been a problem for the scholars to study. The current pain assessment method is mainly based on subjective scores of patients, and comprises the following steps: numerical scoring scales, visual analog scales, language scoring scales, and the like. However, these methods have strong subjective dependencies and cannot form an objective and uniform pain assessment system.
With the rapid development of bioelectricity and imaging technologies such as electroencephalogram, magnetoencephalography and magnetic resonance imaging, objective detection and evaluation of chronic pain through brain activity is attracting attention. For example: DANIEL MARTINS et al found that chronic pain caused remodeling of the structure of the cerebral cortex using a morphological similarity mapping method, and further evaluated the chronic pain; lindquist et al propose a multivariate pattern analysis method for assessing pain in patients. However, the above method does not consider the dynamic change of the brain topological structure in the time dimension and the variability gradient possibly existing between the brain regions, i.e. the abnormal change of chronic pain caused by the macroscopic brain network level cannot be considered, so that the pain degree of the patient is difficult to accurately evaluate.
Disclosure of Invention
The application provides a pain level evaluation method and a pain level evaluation system, which are used for solving the problem that in the prior art, abnormal changes caused by chronic pain on a macroscopic brain network level are not considered, so that the pain level of a patient is difficult to evaluate accurately.
In one aspect, the present application provides a pain class assessment method comprising:
According to the pain sample data, establishing a plurality of first brain networks G 1＝{g1,g2,…,gn corresponding to the pain sample data; establishing a second brain network G 2＝{g1,g2,…,gn corresponding to the control sample data according to the control sample data; establishing a plurality of third brain networks G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn which correspond to the evoked sample data according to the evoked sample data, wherein the third brain network G 3＝{g1,g2,…,gn is obtained according to the data lower than a pain threshold value in the evoked sample data, and the fourth brain network G 4＝{g1,g2,…,gn is obtained according to the data higher than the pain threshold value in the evoked sample data;
Respectively extracting topology attribute values corresponding to the first brain network G 1＝{g1,g2,…,gn, the second brain network G 2＝{g1,g2,…,gn, the third brain network G 3＝{g1,g2,…,gn and the fourth brain network G 4＝{g1,g2,…,gn, wherein the topology attribute values comprise a degree value, a participation coefficient and an aggregation coefficient;
Acquiring a dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, a dynamic variation gradient of the participation coefficient and a dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network G 1＝{g1,g2,…,gn and the topological attribute value of the second brain network G 2＝{g1,g2,…,gn;
acquiring a weight value of each brain region participating in pain perception according to the topological attribute value of the third brain network G 3＝{g1,g2,…,gn and the topological attribute value of the fourth brain network G 4＝{g1,g2,…,gn;
Obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient and the weight value of each brain region participating in pain perception;
and obtaining objective pain grades according to the dynamic pain markers.
Optionally, the establishing a second brain network G 2＝{g1,g2,…,gn corresponding to the control sample data according to the control sample data further includes:
according to the m groups of control sample data, establishing m brain networks corresponding to the m groups of control sample data
Summing the m brain networks and averaging, i.eWherein the method comprises the steps of
Will beDetermined as the second brain network G 2＝{g1,g2,…,gn.
Optionally, the degree value of the first brain network G 1＝{g1,g2,…,gn }, isParticipation coefficientAggregation coefficientThe degree value of the second brain network G 2＝{g1,g2,…,gn }, a degree value of the second brain network G 2＝{g1,g2,…,gn Participation coefficientAggregation coefficient
The obtaining the dynamic variation gradient of the degree value, the dynamic variation gradient of the participation coefficient, and the dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network G 1＝{g1,g2,…,gn and the topological attribute value of the second brain network G 2＝{g1,g2,…,gn, further includes:
According to the degree value of the first brain network G 1＝{g1,g2,…,gn Participation coefficientAggregation coefficientAcquiring corresponding degree value sequence in time dimensionSequence of participation coefficientsAggregation coefficient sequences
According to the degree value of the second brain network G 2＝{g1,g2,…,gn Participation coefficientAggregation coefficientAcquiring corresponding degree value sequence in time dimensionSequence of participation coefficientsAggregation coefficient sequences
According to the degree value sequence corresponding to the first brain network G 1＝{g1,g2,…,gn The participation coefficient sequenceThe aggregation coefficient sequenceThe sequence of degree values corresponding to the second brain network G 1＝{g1,g2,…,gn The participation coefficient sequence The aggregation coefficient sequenceA linear regression model is established to obtain the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn with respect to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, and the dynamic variation gradient of the aggregation coefficient.
Optionally, the establishing a linear regression model includes:
according to the sequence of degree values of the first brain network G 1＝{g1,g2,…,gn And said sequence of degree values of said second brain network G 2＝{g1,g2,…,gn }Establishing a first linear regression model, wherein the first linear regression model is y d＝Kd*Xd +b;
Wherein,
K d is a dynamic variant gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn } relative to the second brain network G 2＝{g1,g2,…,gn.
Optionally, the establishing a linear regression model includes:
According to the coefficient sequence of the first brain network G 1＝{g1,g2,…,gn And the second brain network G 2＝{g1,g2,…,gn }, a sequence of engagement coefficientsEstablishing a second linear regression model, wherein the second linear regression model is y p＝Kp*Xp +b;
Wherein, K p is the dynamic variation gradient of the participation coefficient of the first brain network G 1＝{g1,g2,…,gn } relative to the second brain network G 2＝{g1,g2,…,gn.
Optionally, the establishing a linear regression model includes:
The sequence of aggregation coefficients according to the first brain network G 1＝{g1,g2,…,gn And the aggregate coefficient sequence of the second brain network G 2＝{g1,g2,…,gn }Establishing a third linear regression model, wherein the third linear regression model is y c＝Kc*Xc +b;
Wherein, K c is the dynamic variation gradient of the aggregation coefficient of the first brain network G 1＝{g1,g2,…,gn } relative to the second brain network G 2＝{g1,g2,…,gn.
Optionally, the obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient, and the weight value of each brain region participating in pain perception further includes:
Acquiring a dynamic variation gradient attribute of the first brain network relative to the second brain network according to the dynamic variation gradient K d of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient K p of the participation coefficient and the dynamic variation gradient K c of the aggregation coefficient
And obtaining a dynamic pain marker according to the dynamic variation gradient attribute [ K d,Kp,Kc ] of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn and the weight value of each brain region participating in pain perception.
Optionally, the third brain network G 3＝{g1,g2,…,gn's degree valueParticipation coefficientAggregation coefficientThe degree value of the fourth brain network G 4＝{g1,g2,…,gn }, a degree value of the fourth brain network G 4＝{g1,g2,…,gn Participation coefficientAggregation coefficient
The step of obtaining a weight value of each brain region participating in pain perception according to the topology attribute value of the third brain network G 3＝{g1,g2,…,gn and the topology attribute value of the fourth brain network G 4＝{g1,g2,…,gn, further includes:
acquiring a degree value of G 3＝{g1,g2,…,gn from the third brain network Participation coefficientAnd aggregation coefficientThe obtained topological attribute value matrix of the third brain network
Acquiring a degree value of G 4＝{g1,g2,…,gn from the fourth brain networkParticipation coefficientAggregation coefficientThe obtained topology attribute value matrix of the fourth brain network
According toAndThe method comprises the steps of obtaining a weight value of each brain region participating in pain perception, wherein the weight value alpha of each brain region participating in pain perception is as follows:
α＝([de,pc,cc]w-[de,pc,cc]p)/[de,pc,cc]w。
Optionally, the obtaining objective pain grades according to the dynamic pain markers includes:
Predicting a VAS pain score for the pain sample data based on the dynamic pain markers;
And obtaining objective pain grades according to the VAS pain scores of the predicted pain sample data.
In another aspect, the present application also provides a pain level assessment system configured to:
Establishing a first brain network G 1＝{g1,g2,…,gn from the pain sample data; establishing a second brain network G 2＝{g1,g2,…,gn according to the control sample data; establishing a third brain network G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn according to evoked sample data, wherein the third brain network G 3＝{g1,g2,…,gn is obtained according to data below a pain threshold value in the evoked sample data, and the fourth brain network G 4＝{g1,g2,…,gn is obtained according to data above the pain threshold value in the evoked sample data;
Respectively extracting topology attribute values corresponding to the first brain network G 1＝{g1,g2,…,gn, the second brain network G 2＝{g1,g2,…,gn, the third brain network G 3＝{g1,g2,…,gn and the fourth brain network G 4＝{g1,g2,…,gn, wherein the topology attribute values comprise a degree value, a participation coefficient and an aggregation coefficient;
Acquiring a dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, a dynamic variation gradient of the participation coefficient and a dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network G 1＝{g1,g2,…,gn and the topological attribute value of the second brain network G 2＝{g1,g2,…,gn;
acquiring a weight value of each brain region participating in pain perception according to the topological attribute value of the third brain network G 3＝{g1,g2,…,gn and the topological attribute value of the fourth brain network G 4＝{g1,g2,…,gn;
Obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient and the weight value of each brain region participating in pain perception;
and obtaining objective pain grades according to the dynamic pain markers.
According to the technical scheme, the application provides a pain level evaluation method and a pain level evaluation system, wherein a brain network model corresponding to each sample data is built through three groups of sample data of a pain sample, a control sample and an induction sample, the topological attribute value of each brain network model is calculated, the dynamic variability gradient attribute of the brain network model corresponding to the pain sample data relative to the brain network model corresponding to the control sample data and the weight value of each brain region participating in pain perception are obtained, so that a dynamic pain marker is further obtained, and the perception degree of the brain of a chronic pain patient can be accurately and quantitatively evaluated according to the dynamic pain marker. The method and the system provided by the application can intuitively reflect variability of the chronic pain patient compared with a control group through the dynamic variation gradient attribute of the brain region, and the weight value of each brain region participating in pain perception obtained by inducing a brain network established by sample data can effectively remove pathological influence of the chronic pain, so that the obtained dynamic pain marker is more focused on a somatic perception layer, and a more objective pain grade evaluation result is obtained.
Drawings
Fig. 1 is a flowchart schematically illustrating a pain level assessment method according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to the embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The embodiments described in the examples below do not represent all embodiments consistent with the application. Merely exemplary of systems and methods consistent with aspects of the application as set forth in the claims.
Based on the exemplary embodiments described herein, all other embodiments that may be obtained by one of ordinary skill in the art without making any inventive effort are within the scope of the appended claims. Furthermore, while the present disclosure has been described in terms of an exemplary embodiment or embodiments, it should be understood that each aspect of the disclosure can be practiced separately from the other aspects.
It should be noted that the brief description of the terminology in the present application is for the purpose of facilitating understanding of the embodiments described below only and is not intended to limit the embodiments of the present application. Unless otherwise indicated, these terms should be construed in their ordinary and customary meaning.
The terms first, second, third and the like in the description and in the claims and in the above-described figures are used for distinguishing between similar or similar objects or entities and not necessarily for describing a particular sequential or chronological order, unless otherwise indicated (Unless otherwise indicated). It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are, for example, capable of operation in sequences other than those illustrated or otherwise described herein.
Fig. 1 is a flowchart schematically illustrating a pain level assessment method according to an embodiment of the present application, as shown in fig. 1, where the method includes:
S100: according to the pain sample data, establishing a plurality of first brain networks G 1＝{g1,g2,…,gn corresponding to the pain sample data; establishing a second brain network G 2＝{g1,g2,…,gn corresponding to the control sample data according to the control sample data; establishing a plurality of third brain networks G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn which correspond to the evoked sample data according to the evoked sample data, wherein the third brain network G 3＝{g1,g2,…,gn is obtained according to the data lower than a pain threshold value in the evoked sample data, and the fourth brain network G 4＝{g1,g2,…,gn is obtained according to the data higher than the pain threshold value in the evoked sample data;
The pain sample data are obtained by collecting resting state fMRI images of a plurality of pain patients, the control sample data are obtained by collecting resting state fMRI images of a plurality of healthy people, and the induced sample data are obtained by collecting corresponding task state fMRI images of the healthy people when different temperature stimuli are applied to the healthy people.
In some embodiments, fMRI image data of all pain sample data, control sample data, and evoked sample data may be time-aligned using the SPM12 kit, head-aligned using a rigid transformation to fix the brain in all images to the same target location, then segmenting the brains of each type of sample data into gray matter, white matter, and cerebrospinal fluid and performing a spatial smoothing operation, registering the images to Brain Connectome partition templates (containing 274 brain regions total to brain and cerebellum tissue) and extracting a time series of BOLD signals of 274 brain region gray matter. Using a window of 50 TR times, sliding over the brain region time series of pain sample data and control sample data in 1 TR time step, T-50+1 windows (where T is the number of time points of the brain region time series) can be obtained. And in each window, calculating the Pearson correlation coefficient between every two brain regions, and correcting the Pearson correlation coefficient through Fisher's Z transformation to obtain a z value. The z value of the interval between every two brains is the weight of the node connection of the two brain areas, and the matrix formed by the z values is the brain function connection matrix. The functional connection matrixes corresponding to all windows are arranged according to the window sequence, so that a first brain network G 1＝{g1,g2,…,gn, a second brain network G 2＝{g1,g2,…,gn, a third brain network G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn can be obtained.
S110: and extracting topology attribute values corresponding to all the first brain network G 1＝{g1,g2,…,gn, the second brain network G 2＝{g1,g2,…,gn, the third brain network G 3＝{g1,g2,…,gn and the fourth brain network G 4＝{g1,g2,…,gn, wherein the topology attribute values comprise a degree value, a participation coefficient and an aggregation coefficient.
It should be noted that the degree value of a node in the network is the most direct measure index for describing the importance of the node in the network. The higher the degree value of a node, the more edges there are connected to that node, and the more nodes in the network that are in direct information exchange with that node, and therefore the higher the importance of that node in the network. For an undirected weighted network, the degree value of a node is calculated as follows:
Wherein N represents the total number of nodes in the network, a ij represents the weight of the edge between the node i and the node j, and the weight is equal to the z value obtained by applying Fisher's Z transformation to the corresponding brain region sequence.
The participation factor of a node is a measure that describes the depth of a single node's embedded network local module, expressed by the ratio of the number of nodes' connections within the module to the number of connections within the entire network. The larger the participation coefficient of a node, the more connections the node has in its local module. In the invention, 274 brain areas are divided into 7 cortex sub-networks, subcortical networks and cerebellum networks by adopting functional sub-network division predefined by Yeo et al. The participation coefficients of the nodes are calculated as follows:
where M represents the set of modules in the network, Representing the weight of node i in the network,The weight of the connection of node i to module m is indicated.
The aggregation factor of a node measures the tendency of nodes in a network to aggregate together, expressed by the number of triangles at the node level, as the ratio of the number of edges actually present to the number of edges that are most likely to be present. The aggregate coefficient for a node reflects the likelihood that nodes adjacent to that node are also adjacent, calculated as follows:
Wherein, K i represents the weight of node i in the network, which is the number of triangles around node i.
In some embodiments, the degree value of the first brain network G 1＝{g1,g2,…,gn under the corresponding time node can be obtained according to the above-described manner of solving the degree value, participation coefficient and aggregation coefficientParticipation coefficientAggregation coefficient
In some embodiments, m brain networks corresponding to the m groups of control sample data are establishedSumming the m brain networks and averaging, i.eWherein the method comprises the steps ofWill beDetermined as the second brain network G 2＝{g1,g2,…,gn. And further obtains a degree value of the second brain network G 2＝{g1,g2,…,gn under the corresponding time nodeParticipation coefficientAggregation coefficient
S120: and acquiring a dynamic variation gradient of the degree value, a dynamic variation gradient of the participation coefficient and a dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network and the topological attribute value of the second brain network.
In some embodiments, the degree value according to the first brain network G 1＝{g1,g2,…,gn }, isParticipation coefficient Aggregation coefficientCan obtain the corresponding degree value sequence in the time dimension Sequence of participation coefficientsAggregation coefficient sequences
Degree value according to second brain network G 2＝{g1,g2,…,gn Participation coefficientAggregation coefficient Can obtain the corresponding degree value sequence in the time dimensionSequence of participation coefficientsAggregation coefficient sequences
According to the degree value sequence corresponding to the first brain network G 1＝{g1,g2,…,gn The participation coefficient sequenceThe aggregation coefficient sequenceThe sequence of degree values corresponding to the second brain network G 2＝{g1,g2,…,gn The participation coefficient sequence The aggregation coefficient sequenceAnd establishing a linear regression model to obtain the dynamic variation gradient of the degree value, the dynamic variation gradient of the participation coefficient and the dynamic variation gradient of the aggregation coefficient.
Through the calculation, the brain network topology attribute values of the pain sample data and the control sample data under each time window, namely the time variation sequence of the brain network topology attribute, can be obtained. And calculating to obtain a dynamic variation gradient K d of the brain network topology attribute value by establishing a linear regression model, wherein the dynamic variation gradient K p of the participation coefficient and the dynamic variation gradient K c of the aggregation coefficient are obtained.
In some embodiments, the linear regression model includes a first linear regression model for calculating a dynamic variation gradient of the degree value of the first brain network relative to the second brain network, a second linear regression model for calculating a dynamic variation gradient of the participation coefficient of the first brain network relative to the second brain network, and a third linear regression model for calculating a dynamic variation gradient of the aggregation coefficient of the first brain network relative to the second brain network.
The first linear regression model is y d＝Kd*Xd + b, where, K d is a dynamic variant gradient of the degree value of the first brain network relative to the second brain network.
The second linear regression model is y p＝Kp*Xp + b, where, K p is the dynamic variation gradient of the participation coefficient of the first brain network relative to the second brain network.
The third linear regression model is y c＝Kc*Xc +b, where, K c is the dynamic variation gradient of the aggregation coefficient of the first brain network relative to the second brain network.
In some embodiments, the dynamic variation gradient attribute of the first brain network relative to the second brain network may be further obtained according to the calculated dynamic variation gradient K d of the degree value of the first brain network relative to the second brain network, the dynamic variation gradient K p of the participation coefficient, and the dynamic variation gradient K c of the aggregation coefficientGradient properties according to dynamic variation of a first brain network relative to a second brain networkAnd the weight value of each brain region participating in pain perception, a dynamic pain marker can be obtained, so that the pain level is objectively evaluated.
S130: and acquiring a weight value of each brain region participating in pain perception according to the topological attribute value of the third brain network and the topological attribute value of the fourth brain network.
In some embodiments, the evoked sample data is obtained by acquiring task fMRI images of healthy people when different temperature stimuli are applied, a pain threshold can be preset, a third brain network G 3＝{g1,g2,…,gn is built according to data below the pain threshold in the evoked sample data, a fourth brain network G 4＝{g1,g2,…,gn is built according to data above the pain threshold in the evoked sample data, and the weight of each brain area participating in the pain perception processing process is calculated according to the pain brain network induced by the temperature stimuli, so that the possible influence of chronic pain-induced brain injury on pain perception assessment can be reduced.
In some embodiments, the degree value of the third brain network G 3＝{g1,g2,…,gn Participation coefficientAggregation coefficientDegree value of fourth brain network G 4＝{g1,g2,…,gn Participation coefficientAggregation coefficient
Degree value according to third brain network G 3＝{g1,g2,…,gn Participation coefficientAnd aggregation coefficientA topology attribute value matrix of the third brain network can be obtained
Degree value according to fourth brain network G 4＝{g1,g2,…,gn Participation coefficientAggregation coefficient A topology attribute value matrix of the fourth brain network can be obtained
According toAndThe weight value of each brain region participating in pain perception can be further obtainedThe weight value alpha of each brain region participating in pain perception is as follows:
α＝([de,pc,cc]w-[de,pc,cc]p)/[de,pc,cc]w。
The above calculations may give higher weight to pain management related brain regions while reducing the impact of other region activities on the final assessment.
S140: obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient and the weight value of each brain region participating in pain perception; in some embodiments, the dynamic variant gradient properties obtained above are usedMultiplying the weight value alpha of each brain region participating in pain perception to obtain the dynamic pain marker provided by the application
It should be noted that the dynamic pain marker provided by the application is a noninvasive and sensitive chronic pain biomarker, and can accurately and quantitatively evaluate the perception degree of the brain of a patient suffering from chronic pain to pain.
S150: and obtaining objective pain grades according to the dynamic pain markers.
In some embodiments, the evaluation model used in the present invention is a Lasso regression model, which is a linear model that constructs a penalty function by compression estimation to calculate a refined model. By solving:
arg min||y-Xβ||2s.t.∑βj≤s
a model of the Lasso regression construction can be obtained.
Using the above model, the VAS pain score of pain sample data can be predicted from dynamic pain markers and the effectiveness of this dynamic pain marker assessed using five-fold cross-validation repeated ten times. The validity of the invention is verified by testing that the predicted mean square error of Lasso is 0.04 plus or minus 0.02 (mean plus or minus standard deviation).
According to the above method model, the present application also provides a pain level assessment system configured to:
Establishing a first brain network G 1＝{g1,g2,…,gn from the pain sample data; establishing a second brain network G 2＝{g1,g2,…,gn according to the control sample data; establishing a third brain network G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn according to evoked sample data, wherein the third brain network G 3＝{g1,g2,…,gn is obtained according to data below a pain threshold value in the evoked sample data, and the fourth brain network G 4＝{g1,g2,…,gn is obtained according to data above the pain threshold value in the evoked sample data;
Respectively extracting topology attribute values corresponding to the first brain network G 1＝{g1,g2,…,gn, the second brain network G 2＝{g1,g2,…,gn, the third brain network G 3＝{g1,g2,…,gn and the fourth brain network G 4＝{g1,g2,…,gn, wherein the topology attribute values comprise a degree value, a participation coefficient and an aggregation coefficient;
Acquiring a dynamic variation gradient of the degree value of the first brain network relative to the second brain network, a dynamic variation gradient of the participation coefficient and a dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network and the topological attribute value of the second brain network;
according to the topological attribute value of the third brain network and the topological attribute value of the fourth brain network, obtaining a weight value of each brain region participating in pain perception;
Obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value of the first brain network relative to the second brain network, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient and the weight value of each brain region participating in pain perception;
and obtaining objective pain grades according to the dynamic pain markers.
According to the technical scheme, the application provides a pain level evaluation method and a pain level evaluation system, wherein a brain network model corresponding to each sample data is built through three groups of sample data including a pain sample, a control sample and an induction sample, the topological attribute value of each brain network model is calculated, the dynamic variability gradient attribute of the brain network model corresponding to the pain sample data relative to the brain network model corresponding to the control sample data and the weight value of each brain region participating in pain perception are obtained, so that a dynamic pain marker is further obtained, and the perception degree of the brain of a chronic pain patient can be accurately and quantitatively evaluated according to the dynamic pain marker. According to the method and the system provided by the application, the variability of the brains of the chronic pain patients compared with the variability generated by the control group can be intuitively reflected through the dynamic variation gradient attribute of the brain regions, and the pathological influence of the chronic pain can be effectively removed through the weight value of each brain region participating in pain perception, which is obtained through the brain network established by inducing sample data, so that the obtained dynamic pain marker is focused on the body perception layer, and a more objective pain grade evaluation result is obtained.
In a specific implementation, the present invention further provides a computer storage medium, where the computer storage medium may store a program, where the program may include some or all of the steps in each embodiment of the pain level assessment method and system provided by the present invention when the program is executed. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random-access memory (random access memory RAM), or the like.
It will be apparent to those skilled in the art that the techniques of embodiments of the present invention may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied in essence or what contributes to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present invention.
The same or similar parts between the various embodiments in this specification are referred to each other. In particular, for the display device and the control method embodiment of the display device, since they are substantially similar to the display device embodiment, the description is relatively simple, and the relevant points are referred to the description in the display device embodiment.
The embodiments of the present invention described above do not limit the scope of the present invention.
Claims (10)
1. A method of pain rating assessment, comprising:
According to the pain sample data, establishing a plurality of first brain networks G 1＝{g1,g2,…,gn corresponding to the pain sample data; establishing a second brain network G 2＝{g1,g2,…,gn corresponding to the control sample data according to the control sample data; establishing a plurality of third brain networks G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn which correspond to the evoked sample data according to the evoked sample data, wherein the third brain network G 3＝{g1,g2,…,gn is obtained according to the data lower than a pain threshold value in the evoked sample data, and the fourth brain network G 4＝{g1,g2,…,gn is obtained according to the data higher than the pain threshold value in the evoked sample data;
Extracting topology attribute values corresponding to each of the first brain network G 1＝{g1,g2,…,gn, the second brain network G 2＝{g1,g2,…,gn, the third brain network G 3＝{g1,g2,…,gn and the fourth brain network G 4＝{g1,g2,…,gn, wherein the topology attribute values comprise a degree value, a participation coefficient and an aggregation coefficient;
Wherein, the degree value is calculated as follows:
Wherein N represents the total number of nodes in the network, a ij represents the weight of the edge between the node i and the node j, and the weight is equal to the z value obtained by applying Fisher' sZ transformation to the corresponding brain region time sequence;
The participation coefficients are calculated as follows:
where M represents the set of modules in the network, Representing the weight of node i in the network,The weight of the connection of the node i and the module m is represented;
The aggregation factor is calculated as follows:
Wherein, K i represents the weight of the node i in the network for the number of triangles around the node i;
Acquiring a dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, a dynamic variation gradient of the participation coefficient and a dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network G 1＝{g1,g2,…,gn and the topological attribute value of the second brain network G 2＝{g1,g2,…,gn;
acquiring a weight value of each brain region participating in pain perception according to the topological attribute value of the third brain network G 3＝{g1,g2,…,gn and the topological attribute value of the fourth brain network G 4＝{g1,g2,…,gn;
Obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient and the weight value of each brain region participating in pain perception;
and obtaining objective pain grades according to the dynamic pain markers.
2. The method of claim 1, wherein the establishing a second brain network G 2＝{g1,g2,…,gn corresponding thereto based on the control sample data, further comprises:
according to the m groups of control sample data, establishing m brain networks corresponding to the m groups of control sample data
Summing the m brain networks and averaging, i.eWherein the method comprises the steps of
G' h is determined to be the second brain network G 2＝{g1,g2,…,gn.
3. The method of claim 2, wherein the degree value of the first brain network G 1＝{g1,g2,…,gn } Participation coefficientAggregation coefficientThe degree value of the second brain network G 2＝{g1,g2,…,gn }, a degree value of the second brain network G 2＝{g1,g2,…,gn Participation coefficientAggregation coefficient
The obtaining the dynamic variation gradient of the degree value, the dynamic variation gradient of the participation coefficient, and the dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network G 1＝{g1,g2,…,gn and the topological attribute value of the second brain network G 2＝{g1,g2,…,gn, further includes:
According to the degree value of the first brain network G 1＝{g1,g2,…,gn Participation coefficientAggregation coefficientAcquiring corresponding degree value sequence in time dimensionSequence of participation coefficientsAggregation coefficient sequences
According to the degree value of the second brain network G 2＝{g1,g2,…,gn Participation coefficientAggregation coefficientAcquiring corresponding degree value sequence in time dimensionSequence of participation coefficientsAggregation coefficient sequences
According to the degree value sequence corresponding to the first brain network G 1＝{g1,g2,…,gn The participation coefficient sequenceThe aggregation coefficient sequenceThe sequence of degree values corresponding to the second brain network G 1＝{g1,g2,…,gn The participation coefficient sequence The aggregation coefficient sequenceA linear regression model is established to obtain the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn with respect to the second brain network G 2＝{g1,g2,…,gn}G2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, and the dynamic variation gradient of the aggregation coefficient.
4. A method according to claim 3, wherein said building a linear regression model comprises:
according to the sequence of degree values of the first brain network G 1＝{g1,g2,…,gn And said sequence of degree values of said second brain network G 2＝{g1,g2,…,gn }Establishing a first linear regression model, wherein the first linear regression model is y d＝Kd*Xd +b;
Wherein,
K d is a dynamic variant gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn } relative to the second brain network G 2＝{g1,g2,…,gn.
5. The method of claim 4, wherein the establishing a linear regression model comprises:
a sequence of engagement coefficients according to the first brain network G 1＝{g1,g2,…,gn And the second brain network G 2＝{g1,g2,…,gn }, a sequence of engagement coefficientsEstablishing a second linear regression model, wherein the second linear regression model is y p＝Kp*Xp +b;
Wherein, K p is the dynamic variation gradient of the participation coefficient of the first brain network G 1＝{g1,g2,…,gn } relative to the second brain network G 2＝{g1,g2,…,gn.
6. The method of claim 5, wherein the establishing a linear regression model comprises:
The sequence of aggregation coefficients according to the first brain network G 1＝{g1,g2,…,gn And the aggregate coefficient sequence of the second brain network G 2＝{g1,g2,…,gn }Establishing a third linear regression model, wherein the third linear regression model is y c＝Kc*Xc +b;
Wherein, K c is the dynamic variation gradient of the aggregation coefficient of the first brain network G 1＝{g1,g2,…,gn } relative to the second brain network G 2＝{g1,g2,…,gn.
7. The method of claim 6, wherein the deriving a dynamic pain marker from the dynamic variability gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variability gradient of the participation coefficient, the dynamic variability gradient of the aggregation coefficient, and the weighting value of the participation pain perception in each brain region further comprises:
Acquiring a dynamic variation gradient attribute of the first brain network relative to the second brain network according to the dynamic variation gradient K d of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient K p of the participation coefficient and the dynamic variation gradient K c of the aggregation coefficient
And obtaining a dynamic pain marker according to the dynamic variation gradient attribute [ K d,Kp,Kc ] of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn and the weight value of each brain region participating in pain perception.
8. The method of claim 7, wherein the third brain network G 3＝{g1,g2,…,gn's metric value Participation coefficientAggregation coefficientThe degree value of the fourth brain network G 4＝{g1,g2,…,gn }, a degree value of the fourth brain network G 4＝{g1,g2,…,gn Participation coefficientAggregation coefficient
The step of obtaining a weight value of each brain region participating in pain perception according to the topology attribute value of the third brain network G 3＝{g1,g2,…,gn and the topology attribute value of the fourth brain network G 4＝{g1,g2,…,gn, further includes:
acquiring a degree value of G 3＝{g1,g2,…,gn from the third brain network Participation coefficientAnd aggregation coefficientThe obtained topological attribute value matrix of the third brain network
Acquiring a degree value of G 4＝{g1,g2,…,gn from the fourth brain networkParticipation coefficientAggregation coefficientThe obtained topology attribute value matrix of the fourth brain network
According toAndThe method comprises the steps of obtaining a weight value of each brain region participating in pain perception, wherein the weight value alpha of each brain region participating in pain perception is as follows:
α＝[de,pc,cc]w-[de,pc,cc]p/[de,pc,cc]w。
9. the method of claim 1, wherein said deriving an objective pain rating from said dynamic pain marker comprises:
Predicting a VAS pain score for the pain sample data based on the dynamic pain markers;
And obtaining objective pain grades according to the VAS pain scores of the predicted pain sample data.
10. A pain level assessment system, the system configured to:
Establishing a first brain network G 1＝{g1,g2,…,gn from the pain sample data; establishing a second brain network G 2＝{g1,g2,…,gn according to the control sample data; establishing a third brain network G 3＝{g1,g2,…,gn and a fourth brain network G 4＝{g1,g2,…,gn according to evoked sample data, wherein the third brain network G 3＝{g1,g2,…,gn is obtained according to data below a pain threshold value in the evoked sample data, and the fourth brain network G 4＝{g1,g2,…,gn is obtained according to data above the pain threshold value in the evoked sample data;
Respectively extracting topology attribute values corresponding to the first brain network G 1＝{g1,g2,…,gn, the second brain network G 2＝{g1,g2,…,gn, the third brain network G 3＝{g1,g2,…,gn and the fourth brain network G 4＝{g1,g2,…,gn, wherein the topology attribute values comprise a degree value, a participation coefficient and an aggregation coefficient;
Wherein, the degree value is calculated as follows:
Wherein N represents the total number of nodes in the network, a ij represents the weight of the edge between the node i and the node j, and the weight is equal to the z value obtained by applying Fisher' sZ transformation to the corresponding brain region time sequence;
The participation coefficients are calculated as follows:
where M represents the set of modules in the network, Representing the weight of node i in the network,The weight of the connection of the node i and the module m is represented;
The aggregation factor is calculated as follows:
Wherein, K i represents the weight of the node i in the network for the number of triangles around the node i;
Acquiring a dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, a dynamic variation gradient of the participation coefficient and a dynamic variation gradient of the aggregation coefficient according to the topological attribute value of the first brain network G 1＝{g1,g2,…,gn and the topological attribute value of the second brain network G 2＝{g1,g2,…,gn;
Acquiring a weight value of each brain region participating in pain perception according to the topological attribute value of the third brain network G 3＝{g1,g2,…,gn and the topological attribute value of the fourth brain network G 4＝{g1,g2,…,gn;
Obtaining a dynamic pain marker according to the dynamic variation gradient of the degree value of the first brain network G 1＝{g1,g2,…,gn relative to the second brain network G 2＝{g1,g2,…,gn, the dynamic variation gradient of the participation coefficient, the dynamic variation gradient of the aggregation coefficient and the weight value of each brain region participating in pain perception;
and obtaining objective pain grades according to the dynamic pain markers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111582180.5A CN114283935B (en) | 2021-12-22 | 2021-12-22 | Pain grade assessment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111582180.5A CN114283935B (en) | 2021-12-22 | 2021-12-22 | Pain grade assessment method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114283935A CN114283935A (en) | 2022-04-05 |
| CN114283935B true CN114283935B (en) | 2024-07-16 |
Family
ID=80873868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111582180.5A Active CN114283935B (en) | 2021-12-22 | 2021-12-22 | Pain grade assessment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114283935B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102266223A (en) * | 2010-06-01 | 2011-12-07 | 四川大学华西医院 | Pain assessment system based on magnetic resonance resting state functional imaging |
| CN104715150A (en) * | 2015-03-19 | 2015-06-17 | 上海海事大学 | Migraineur cerebral cortex assistant classification analyzing method based on complex network |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140066739A1 (en) * | 2012-08-29 | 2014-03-06 | Bin He | System and method for quantifying or imaging pain using electrophysiological measurements |
| CN111951958B (en) * | 2020-08-21 | 2024-07-12 | 深圳大学 | Pain data evaluation method based on self-coding and related components |
-
2021
- 2021-12-22 CN CN202111582180.5A patent/CN114283935B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102266223A (en) * | 2010-06-01 | 2011-12-07 | 四川大学华西医院 | Pain assessment system based on magnetic resonance resting state functional imaging |
| CN104715150A (en) * | 2015-03-19 | 2015-06-17 | 上海海事大学 | Migraineur cerebral cortex assistant classification analyzing method based on complex network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114283935A (en) | 2022-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Proix et al. | Individual brain structure and modelling predict seizure propagation | |
| Rocha et al. | Recovery of neural dynamics criticality in personalized whole-brain models of stroke | |
| CN115995299B (en) | Method, system, electronic equipment and medium for positioning individual regulation and control targets | |
| CN111739647B (en) | Magnetic resonance image-based bipolar disorder suicide risk prediction method and related device | |
| CN110322554B (en) | Brain dynamic function mode stability calculation method | |
| JP2002534144A (en) | Method and apparatus for performing a visual field test and computer program for processing the result | |
| Morgan et al. | Integrating functional and diffusion magnetic resonance imaging for analysis of structure-function relationship in the human language network | |
| Martinez-Vargas et al. | Improved localization of seizure onset zones using spatiotemporal constraints and time-varying source connectivity | |
| US20220262515A1 (en) | Method for determining an onset time and an excitability of a brain region | |
| CN110326054A (en) | Determine the method, apparatus and program for participating in executing at least one brain network of given process | |
| CN115337000B (en) | Machine learning method for evaluating brain aging caused by diseases based on brain structure images | |
| CN114970599A (en) | Identification method and identification device for attention defect associated electroencephalogram signals and storage medium | |
| CN112614126A (en) | Magnetic resonance image brain region dividing method, system and device based on machine learning | |
| Dai et al. | Predicting individual brain functional connectivity using a Bayesian hierarchical model | |
| CN107785079B (en) | Depression patient disease recovery assessment method based on diffusion tensor imaging | |
| CN114708263B (en) | Individual brain functional region positioning method, device, equipment and storage medium | |
| Barnes et al. | Individual differences in intrinsic brain connectivity predict decision strategy | |
| Giulia et al. | The impact of neurofeedback on effective connectivity networks in chronic stroke patients: an exploratory study | |
| Nandakumar et al. | DeepEZ: a graph convolutional network for automated epileptogenic zone localization from resting-state fMRI connectivity | |
| Gelskov et al. | Amygdala signals subjective appetitiveness and aversiveness of mixed gambles | |
| CN114283935B (en) | Pain grade assessment method and system | |
| Wilmskoetter et al. | Dynamic network properties of the superior temporal gyrus mediate the impact of brain age gap on chronic aphasia severity | |
| CN111938671A (en) | Anxiety trait quantification method based on multi-dimensional internal perception features | |
| Olsbo et al. | Development and evaluation of spatial point process models for epidermal nerve fibers | |
| CN113827256B (en) | Fatigue detection method, device, terminal and storage medium based on brain electrical micro state |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |